Dynamic Shielding for Reinforcement Learning in Black-Box Environments.
Masaki WagaEzequiel CastellanoSasinee PruekprasertStefan KlikovitsToru TakisakaIchiro HasuoPublished in: ATVA (2022)
Keyphrases
- black box
- reinforcement learning
- dynamic environments
- black boxes
- white box
- integration testing
- highly dynamic
- rule extraction
- white box testing
- artificial intelligence
- hybrid systems
- test cases
- function approximation
- learning algorithm
- database
- software development
- state space
- dynamic programming
- multi agent
- neural network